Administrators can configure an acceleration engine. If reports are created based on MaxCompute data sources, you can configure a Data Lake Analytics (DLA) acceleration engine in the Quick BI console and use DLA to accelerate queries on the MaxCompute data sources. This reduces report query time and improves user experience.

Prerequisites

The MaxCompute data source is connected. For more information, see Add a cloud data source MaxCompute.

Limits

  • Only DLA Serverless Presto virtual clusters can be used to accelerate queries on MaxCompute data sources.
  • The DLA Serverless Presto virtual cluster that is used must be in the same region as the MaxCompute project to which the data source belongs.

Precautions

When you create a DLA Serverless Presto virtual cluster of DLA CU Edition, we recommend that you set the Long-term retained resources (MIN) parameter to a specification of 32 cores and 128 GB memory or higher. If you select a specification lower than 32 cores and 128 GB memory, the system cannot provide high acceleration performance.

Step 1: Create a DLA Serverless Presto virtual cluster of DLA CU Edition

Only DLA Serverless Presto virtual clusters of DLA CU Edition provide the query acceleration feature for MaxCompute data sources. You must create a DLA Serverless Presto virtual cluster of DLA CU Edition. For more information, see Create a virtual cluster
Notice
  • Only DLA Serverless Presto virtual clusters can be used to accelerate queries on MaxCompute data sources.
  • The DLA Serverless Presto virtual cluster that is used must be in the same region as the MaxCompute project to which the data source belongs.
  • We recommend that you set the long-term retained resources (MIN) parameter to a specification of 32 cores and 128 GB memory or higher. If you select a specification lower than 32 cores and 128 GB memory, the system cannot provide high acceleration performance.

Step 2: Configure a whitelist for the DLA virtual cluster

Before you configure an acceleration engine in the Quick BI console, add the CIDR blocks 10.152.69.0/24, 10.152.163.0/24, and 139.224.4.0/24 to the IP address whitelist of the DLA virtual cluster. If you do not add these CIDR blocks to the whitelist of the DLA virtual cluster, Quick BI cannot connect to the DLA virtual cluster and queries cannot be accelerated. For information about how to configure a DLA whitelist, see Create an endpoint.

Step 3: Configure a DLA acceleration engine

  1. In the top navigation bar, click the Settings icon.
  2. In the left-side navigation pane, click System Configuration.
  3. On the System Configuration page, click the Acceleration engine tab.
    Acceleration engine
  4. Configure the following parameters and click save.
    Parameter Description
    Acceleration engine type Only DLA Serverless Presto engine is supported.
    Database address The public endpoint of DLA.

    You can go to the SQL access point page in the DLA console to view the public endpoint. For more information, see Create a public endpoint.

    port The public port of DLA.

    You can go to the SQL access point page in the DLA console to view the public port. For more information, see Create a public endpoint.

    username The account that is used to connect to DLA.

    You can view your DLA account name on the Account page in the DLA console. For more information, see Create a database sub-account.

    Note After you activate DLA, the system automatically creates a root account. You can enter the root account in the username field. If you want to use a RAM user, make sure that the RAM user has full permissions on DLA.
    password The password that is used to connect to DLA.

    You can reset the password.

    virtual cluster The name of the DLA virtual cluster.

    You can obtain the cluster name on the Virtual Cluster Management page in the DLA console. For more information, see Create a virtual cluster.

After you configure the acceleration engine, you can enable and use the acceleration feature for datasets. For more information, see Configure cache acceleration. Acceleration and cache